Discrete emotions discovered by contactless measurement of facial blood flows

Abstract

Experiential and behavioural aspects of emotions can be measured readily but developing a contactless measure of emotions’ physiological aspects has been a major challenge. We hypothesised that different emotion-evoking films can produce distinctive facial blood flow patterns that can serve as physiological signatures of discrete emotions. To test this hypothesis, we created a new Transdermal Optical Imaging system that uses a conventional video camera to capture facial blood flows in a contactless manner. Using this and deep machine learning, we analysed videos of the faces of people as they viewed film clips that elicited joy, sadness, disgust, fear or a neutral state. We found that each of these elicited a distinct blood flow pattern in the facial epidermis, and that Transdermal Optical Imaging is an effective contactless and inexpensive tool to the reveal physiological correlates of discrete emotions.